Ed Bayes, a designer on the Codex team, offered some insight into how this tool has transformed his workflow. Codex now connects seamlessly with project management tools like Linear and communication platforms like Slack. This means team members can assign coding tasks directly to the AI. “You can add Codex, and you can assign issues to Codex,” he shared with Ars. “Codex acts like a teammate in your workspace.”
This new integration allows users to tag Codex in Slack when they have feedback. For example, if there’s an issue, they can ask Codex to fix it. The AI then creates a pull request, and the team can discuss changes right in the Slack thread. “It’s like having a coworker show up wherever you are,” Bayes noted.
For Bayes, who focuses on visual design and interactions, Codex has made it possible for him to write code directly. He doesn’t just pass specifications to engineers anymore. “It gives you more leverage,” he said. Designers at OpenAI now prototype features by building them with Codex handling the tricky details.
OpenAI views Codex as a “junior developer,” with plans for it to progress to a senior level over time. Bayes emphasized this point: “If you were onboarding a junior developer, you’d give them a Slack account and a Linear account.” Codex isn’t just a tool; it’s a part of the team.
But what does this mean for human developers? There’s some debate here. Embiricos explains the difference between “vibe coding”—where developers trust AI-generated code without much scrutiny—and “vibe engineering,” a term coined by AI researcher Simon Willison. In vibe engineering, humans are still involved, iterating on the AI’s suggestions and reviewing the code carefully. “We see a lot more vibe engineering in our code base,” Embiricos stated. “You might ask Codex for a plan first, collaborate on that, then make sure to review the outcome.”
As of 2023, studies show that many developers feel AI tools enhance their productivity rather than replace their work. A recent survey from Stack Overflow revealed that over 60% of developers now use some form of AI assistance in their coding tasks.
With Codex, users are not just passive consumers of technology; they’re active participants in a collaborative process that could reshape the future of development. As AI continues to evolve, the balance between human insight and machine efficiency will be critical in ensuring quality and creativity in coding.
For more details on how AI is shaping the tech landscape, you can check out reports from reputable sources like Pew Research Center.
